Equalizing loudspeakers based on anechoic measurements (community project)

[dasdoing]

Given that there's always a cost/penalty in time, how low a delay (ms) were you able to achieve?

whole DSP delay is 229ms. this includes the filter, the LP software crossover and system delay [EDIT] and 5,1ms delay to the mains

 

[dasdoing]

Given that there's always a cost/penalty in time, how low a delay (ms) were you able to achieve?

I was actualy thinking of you yesterday cause I would like to try my new concept/DRC settings aprouch in a "normal room" to see how close the resulting in room curve is to Harman. I still have that delay test mdat of yours

 

[ernestcarl]

I was actualy thinking of you yesterday cause I would like to try my new concept/DRC settings aprouch in a "normal room" to see how close the resulting in room curve is to Harman. I still have that delay test mdat of yours

Don't quite remember... which test?

 

[dasdoing]

Don't quite remember... which test?

how low a delay (ms) were you able to achieve?

like someone in a diferent forum once wrote: I just press play a little faster lol

I know, the issue is with movies, but my movie play lets me delay the video against the sound

 

[Pio2001]

I feel like that when we acept that we hear through the room above 1000Hz-ish, the in-room graph in that range is then irelevant.

I think that both in-room and direct sound are equally relevant.

this is a linear phase correction done with http://drc-fir.sourceforge.net/
this is the flattest I can get with this software because equalizing those 2 dips caused by the crossovers further would result in pre-delay (fun fact: they disapear in the normal curve).

Use minimum phase to avoid pre-echo.

 

[dasdoing]

I think that both in-room and direct sound are equally relevant.

I wonder if recording music and/or pink noise through the speakers and then ABing it through headphones could answer this doubt

Use minimum phase to avoid pre-echo.

I use mixed phase. I showed here that some of the dips require phase correction: https://www.audiosciencereview.com/...stortion-abx-testing.18709/page-2#post-657930

 

[Pio2001]

I wonder if recording music and/or pink noise through the speakers and then ABing it through headphones could answer this doubt

Not with omnidirectional mics. Some of the information that allow our ears to distinguish direct and diffuse sound would be lost in the recording.

But a binaural recording with an artificial head, played back with earbuds, could work.

 

[ernestcarl]

I know, the issue is with movies, but my movie play lets me delay the video against the sound

I have a vague feeling we had an exchange about something along the those lines maybe a few months before(?), but I still can't quite remember what the specific topic or thread was. Sorry!

With regards to your convolution, the delay is still a little bit too much for me personally with regards to direct streaming videos e.g. Crave, Netflix, Youtube, Disney+ etc. This is not a problem with physical discs and locally downloaded media, of course.

 

[dasdoing]

I have a vague feeling we had an exchange about something along the those lines maybe a few months before(?), but I still can't quite remember what the specific topic or thread was. Sorry!

it was in the subwoofer time alignment topic

With regards to your convolution, the delay is still a little bit too much for me personally with regards to direct streaming videos e.g. Crave, Netflix, Youtube, Disney+ etc. This is not a problem with physical discs and locally downloaded media, of course.

software outside the OS is a problem, yea. my TV can only aply positive delay to the audio.
for Youtube inside the OS there is a solution for example: https://chrome.google.com/webstore/detail/youtube-audiovideo-sync/mknmhikmjljhpccebpnplhicmcfjkgbk

 

[ernestcarl]

it was in the subwoofer time alignment topic



software outside the OS is a problem, yea. my TV can only aply positive delay to the audio.
for Youtube inside the OS there is a solution for example: https://chrome.google.com/webstore/detail/youtube-audiovideo-sync/mknmhikmjljhpccebpnplhicmcfjkgbk

Haven't seen that app/extension solution before. Could work if the only streaming platform I used was Youtube. Yeah... I still use the same miniDSP 2x4HD -- and nearly filled all PEQ bands with simple IIR EQ. Also tried using its FIR feature, but I wasn't really satisfied with the lack of filter taps and delay. Since I already use JRiver, it was more expedient to just use its WDM driver and route to it directly.

 

[dasdoing]

Haven't seen that app/extension solution before. Could work if the only streaming platform I used was Youtube. Yeah... I still use the same miniDSP 2x4HD -- and nearly filled all PEQ bands with simple IIR EQ. Also tried using its FIR feature, but I wasn't really satisfied with the lack of filter taps and delay. Since I already use JRiver, it was more expedient to just use its WDM driver and route to it directly.

yea, MiniDsp is not realy usable for serious FIR usage

 

[aarons915]

Some may know I bought some Neumann KH 120 a few weeks ago to try them out in my home setup. I wanted to like them and they were very good speakers but unfortunately they were tiring to listen to and had to return them but I did learn quite a bit in the process and wanted to share. This is one example where a CTA-2034 measurement by itself isn't enough info and the separate horizontal and vertical graphs are useful. If you check out the MMM around my listening position it seems there is a large peak at 3k that would be responsible for my listening fatigue:



But I tried smoothing out that bump and it did help but not enough. Also, if we check the Spin, horizontal and vertical measurements posted by Neumann in the KH80 thread, you can't really EQ much around 3k without messing up the near perfect horizontal response.



You can see that the real reason that there appears to be a peak around 3k in my in-room response is due to the large dips at 2 and 4k in the vertical plane. What I think is happening is the direct sound and sidewall reflections are all very close and in agreement but the radically different ceiling and floor reflections are confusing to the auditory system and causing fatigue. There are no peaks anywhere in the response so I'm not sure what else it could be but if someone has another theory I'd be interested in hearing it.

Of course I'm not criticizing these speakers, they are designed for near field use where this would never be an issue, these sounded great in the short term and I preferred them over my KEF R3 in a direct A/B comparison. They are also very solid speakers and I really like the build quality, I was bummed that I couldn't make them work in a home setup.

 

[Chromatischism]

It's possible that in their intended application of near-field monitoring, there is a very small emphasis in that 3k range to clearly hear the details of your mix. Many studio monitors have a similar feature.

 

[dasdoing]

I see no 3k peak.
I see a broadband 5dB lack below 200Hz-ish which will obviously result in very harsh mids

 

[aarons915]

It's possible that in their intended application of near-field monitoring, there is a very small emphasis in that 3k range to clearly hear the details of your mix. Many studio monitors have a similar feature.

I see no 3k peak.
I see a broadband 5dB lack below 200Hz-ish which will obviously result in very harsh mids

Agreed, maybe a slight 3k bump in the listening window but I EQ'd the response flat after 1k and the fatigue still happened. Regarding my in room response, it does look bass shy but this is each speaker measured by itself so when you have 2 together the bass is higher in level but I don't like as much bass as most people so mine is still flat without a house curve that some people like. Either way, that can't be responsible because my bass is always consistent and other speakers I've used haven't been fatiguing.

 

[Presently42]

I would like to participate in testing, compiling and provide my own take on eq. I have the JBL A130, acquired after a glowing review by our host. I've tried @Maiky76's two eq profiles, as well as those of a few other members - six different eq profiles, all told; as well as one I created myself. Maiky76's listening window eq profile was my go to, until I tried my own, which I now seem to prefer. I do seemingly occasionally prefer the score-optimised one. The differences between these three are small, but sufficient, that I feel its inclusion in this debate to be worthwhile. As I'm lazy, I'll quote the post I made in the A130 thread:

I'm trying currently another set of filters, this time based on Floyd Toole's research, presented in the graph in this post. It's the winner so far. I averaged the on axis and early reflections curves, and used those to create filters with only negative values between Schroeder and 5 kHz, with a -0.2 dB / oct slope; then used flat on axis for everything above. I justify this as follows:

1. The graph indicated, that the main sonic components between Schroeder and a bit under 5 kHz are an amalgamaish of on axis and early reflections; with sound power providing marginal effects. I might play with attempting to include sound power; having done this once, and having not terribly liked the results, I have chosen now to ignore it. We'll see.
2. Seemingly, research has shown, that a slope of -0.2 dB/oct is preferred for the listening window. Given, that the listening window is very similar to the average of the on axis and early reflections, I chose to keep this value. I cannot currently find the proof of this. Nevertheless, the results sound good. Further investigaish in to this tilt is warranted.1
3. The graph indicates, that above roughly 5 kHz, the on axis is the dominant sound field producer.
4. Given, that research indicates a preference for flat on axis response, I've not put a tilt to this on axis curve.

Lastly, in the final eq profile, I've included a tilt, starting at 1.3 kHz (a value obtained from more than one Harman graphs. I briefly tried applying tilts starting at different values, and preferred the one starting at 1.3 kHz. However, I'm not convinced; and should like to investigate further.), of -0.6 dB/oct; figuring, that this tilt, with the tilt of -0.2 dB/oct already (somewhat) present, makes for a tilt of -0.8 - which I seem to prefer in room, and which one on-line calculator suggested; and which at least one other member agrees, is a fine amount of tilt. That being said, this tilt is room, programme and mood dependant; like salt, I add to taste.2

I'll take some measurements later and see exactly what's going on. Tbh, the differences are quite subtle; and I might prefer this new one merely because it is new. Measurements and more blind testing are yet needed!3

The blue curve is Maiky76's score optimised one; the orange on is the listening window one without the filters below 400 Hz, nor the one around 5 kHz; and the teal one is my own. REW's variable smoothing is used. The trough at 1.8 kHz is the only real flaw; but as it's present in all three (I think it's also there without any eq), I'm not sure what's causing it, and I'm still investigating it.

1: I investigated the matter of tilt a bit; and with only using the on axis, as opposed to an average of on axis and early reflections. The results were so similar, that the quick abx testing I did yielded no conclusive results.
2: Turns out, based on the setup instructions of Equaliser APO, I had enabled the Windows function called Enable audio enhancements, which was putting a hump at around 4.4 kHz and 10 kHz. Much annoy. The speakers now exhibit a natural tilt of around -0.5 db / oct, a fine amount that Toole seems to suggest, in one graph, is the preferred amount for a good speaker in a good room. I'm still figuring out the full effect this Enable audio enhancements function had on my system, so this part might change. And anyway, it's programme (and mood!) dependant and whatnot!
3: Indeed, my eq profile is essentially identical to the listening window one, other than a greater subduing of frequencies above around 10 kHz.

 

[Presently42]

So as to not edit the above post again, permit me to create this new one.

I observed a few things, tested their fixes and measured the effects:




Flat as you like. But how have we achieved this? Recall what Toole has said: a speaker's resonances can be fixed - but nothing much else; and, that these resonances can be observed as peeks in the on-axis, early reflections and sound power curves. As can be noted in the mmm curves in my previous post, there was a trough at around 1.8 kHz - as well as a spectacular falling off around 15 kHz or so. We'll note, that both of these areas are ones of resonance in the on-axis, early reflections and listening window curves - but not in the sound power curve (well, the one at 15 kHz is - but we'll return to that later).

I continued using the averaging of the on-axis and listening window curves with a 0.2 dB/oct tilt - but this time, I looked at the sound power curve. Putting a line through it (a tilt of 1.3 dB/oct, though I doubt this number is significant), we notice, that the supposed resonance at around 1.8 kHz has disappeared:



I therefore instructed REW to create negative filters on the on-axis listening window average curve between 400 Hz (where the sound power starts to have less of an effect; which is also seemingly the Schroeder frequency) and about 1.6 kHz, where the sound power curve meets the tilt. This created a spectacular set of three filters, and led to the in-room responses seen above, which are unexpectedly flat above Schroeder. But there's something missing: the response above ~12 kHz is still a bit wonky.

As noted by Toole, the on-axis curve is mostly responsible for the in-room sound above ~5 kHz. To this effect, I created a set of filters, whose effects were thus:



Essentially, we're listening to the filters themselves. This interesting filter audibility is noted by member @Pio2001 in his own set of experiments. In this red curve above, I instructed REW to create filters on th on-axis curve between 5 kHz, roughly the frequency which Toole notes delineates beginning of the the on-axis dominated sound field; and just before that tremendous spike. The spike itself I choose to not eq at all. The results, in the red curve, are flat until a small 2 dB peak, which is obviously the spike - but at much reduced volume. Quite honestly, I cannot yet determine what effects the spike is having, nor how to predict them - nor even how to compensate for them. The investigation continues!

To conclude, my findings are these:

1. An average of the on-axis and early reflections curves, with a tilt of 0.2 dB/oct, seems to yield excellent results between Schroeder (~400 Hz) and ~5 kHz.
2. The on-axis curve with no tilt is to be used above ~5 kHz.
3. Peaks present in all of the spin curves should be eq'd. If a peak is not present in one of the curves, especially the sound power curve, the peak must not be eq'd, no matter how ugly that makes the resulting on-axis early reflections average curve.
4. Only negative filters should be applied. (I'm not yet convinced of this, but it agrees with my findings and those of Toole. That being said, the A130 doesn't need boosting anywhere, so this point is not very strong.)
5. Filters above ~9 kHz are individually heard. (This is a fairly arbitrary number: more testing is needed.)

Other speakers must be tested to see if this holds true for them as well. If so, however, automatically generated eq filters will be most difficult, as all curves must be consulted in order to ensure, that the proposed resonance is indeed a resonance and not something else, which cannot - or at least should not - be eq'd.

Are there any here who'd be willing to test a set of filters based on these five principles? I'd love to see how strongly they hold on other speakers.

Edit: I forgot to mention my subjective evaluation! I listened to a solid ten seconds of Fast Car: it sounded very smooth indeed. Of course, further listening is required - but I'm hungry, so it's time to cook.

 

[Robbo99999]

What seems to work well is EQ'ing for a flat(tish) listening window, subsequently applying the filters to the on-axis response, and using the appropriate shelf to make sure there is no excess energy in the on-axis curve. This way the narrow band EQ filters are only applied on the listening window average and eliminates the chance that you are EQ'ing on-axis interference effects which are not meant to be EQ'd.

When this is done the filters can be applied to the early reflections (or in-room estimate) curve and see if nothing too weird is going there. Faults of omission are generally considered more benign than faults of excess.

Here's my F206s

View attachment 75583
View attachment 75584

Hi Tim, I'm revisiting my Anechoic LW EQ of my JBL 308p Mkii speakers. I took your advice and came up with the following for a flat Listening Window:

I then applied the same filters to the on-axis frequency response which is the following:

How does that look?? I was a bit bothered that there might be some excess energy being shown in on-axis above 1kHz (on average)? So that prompted me to apply a Tone Control Filter to bring the On-Axis to a flatter tonal response, which is Filter #7 in the following pic (which basically puts in an almost linear tilt from 100Hz up to 20000Hz):

Would you say that Tone Control Filter (Filter#7) has made an improvement to the on-axis?
And here's the effect of that Tone Control Filter when applied to the Listening Window:

The Tone Control Filter makes a noticeable difference when listening, but I'm on the fence about whether I actually like it more or less, it varies. What do you think, have you got any advice for me re what you're seeing here (open to advice from other people too, not just the OP)?

 

[reficver]

Very interesting!
Your EQ for the 8030C, 20 EQ points with many tiny sharp adjustments towards NBD: 6.9

My EQ 5 points: score 6.83
PEQ 273.7Hz, 1.17dB, 0.94,...
PEQ 1720.0Hz, -2.27dB, 2.52,...
PEQ 2250.0Hz, 2.25dB, 0.88,...
PEQ 4630.0Hz, -0.97dB, 7.30,...
PEQ 6785.0Hz, -0.73dB, 4.44,...

The NBD is just not the driving factor as I explained, the EQ I design are based on diminishing return based directly on the score.

Hello Maiky, Thanks for Your EQing.

Your recent EQ (maybe after 21 years?) seems to be shared by dividing it into EQ LW and EQ SCORE, unlike the old threads.
Could you please write EQ LW/EQ SCORE for Genelec 8030c one more time with recent algorithm? Thank you.

 

[Igor Kirkwood]

I think that both in-room and direct sound are equally relevant.



Use minimum phase to avoid pre-echo.

It is perfectly possible to correct in linear phase FIR, with a "pre echo" almost zero (- 70 dB) as indicated by this response curve by Jean-Luc Ohl, French Engineer Measurer